ABSTRACT There is growing evidence that breast cancer patients with adverse body composition (low muscle mass and excess adiposity) have worse therapeutic responses and an increased risk of recurrence and death. Both low muscle mass and excess adiposity promote systemic metabolic abnormalities like the metabolic syndrome. While the impact of adverse body composition on the local tumor microenvironment remains unclear, data from pre-clinical animal models and in vitro analyses suggest that adverse body composition could contribute to impaired anti-tumor immunity and alterations in metabolic and proliferative signaling pathways. However, these relationships have not been comprehensively studied in human breast cancer patients or within molecular subtypes of breast cancer that have differing biology and prognosis. Developing and personalizing breast cancer therapies will require understanding how systemic factors such as patient body composition and the metabolic syndrome impact the local breast tumor microenvironment and therefore breast cancer survival. To address this gap in knowledge, we propose a molecular epidemiologic study to determine if lower muscle mass, excess adiposity and the metabolic syndrome impair the anti-tumor immune response by promoting T- cell exhaustion (Aim 1) and/or by altering the activity of PI3K/AKT/mTOR pathways and proliferative signaling (Aim 2) in the breast tumor microenvironment. We will further determine if alterations in these immune, metabolic and proliferative signaling pathways mediate the association of adverse body composition with breast cancer recurrence and breast cancer-specific mortality (Aim 3). To accomplish this, we will measure gene expression levels in 1,400 archived clinical breast tumor samples selected from a unique cohort: more than 4,000 stage II and III breast cancer patients with longitudinal follow-up and precise measures of muscle, subcutaneous, and visceral adipose tissue. All these patients also have rich electronic medical record data on cancer treatments and the metabolic syndrome. Since our sampling approach ensures representation across each of four major clinicopathological categories defined by hormone receptor and human epidermal growth factor receptor 2 status, we will be able to examine associations overall, and separately within each breast cancer subtype. Using this premier data resource, we will examine how differences in the breast tumor microenvironment caused by adverse body composition and the metabolic syndrome may mediate differences in long-term breast cancer outcomes. The proposed study will help personalize existing breast cancer therapies and develop future intervention approaches that consider both the molecular features of the tumor and patient body composition. Furthermore, this study will identify tumor biomarkers that are appropriate targets to be measured in future trials to improve body composition and the metabolic syndrome through pharmacologic or lifestyle intervention.